Subgroup analysis of unweighted ORR difference per BICR in all randomly assigned patients. Two-sided 95% CI for unweighted difference was calculated using Newcombe method. ORR difference is not computed for subsets with <10 patients in at least one treatment group. ORR difference: NIVO plus RELA minus NIVO. aORR difference was not computed for subset category with <10 patients in at least one treatment group. BICR, blinded independent central review; chemo, chemotherapy; CPS, combined positive score; ECOG PS, Eastern Cooperative Oncology Group performance status; FOLFOX, oxaliplatin + leucovorin + fluorouracil; GC, gastric cancer; GEJC, gastroesophageal junction cancer; LAG-3, lymphocyte-activation gene 3; NIVO, nivolumab; ORR, objective response rate; RELA, relatlimab; XELOX, oxaliplatin + capecitabine.