Pembrolizumab plus FLOT as neoadjuvant and adjuvant therapy in locally advanced gastric and gastroesophageal junction cancer: analysis of the FLOT cohort from the phase 3 KEYNOTE-585 study

BACKGROUND: Results from the third interim analysis of the phase 3 KEYNOTE-585 study (NCT03221426) showed superior pathological complete response (pCR) with perioperative pembrolizumab and chemotherapy followed by adjuvant pembrolizumab versus placebo plus chemotherapy, then adjuvant placebo in locally advanced, resectable gastric or gastroesophageal junction adenocarcinoma; differences in event-free survival (EFS) were not statistically significant. We present results from a complementary cohort that evaluated pembrolizumab plus fluorouracil, leucovorin, oxaliplatin, and docetaxel (FLOT) versus placebo plus FLOT. PATIENTS AND METHODS: KEYNOTE-585 is a randomized, phase 3 study conducted at 143 medical centers. Participants were randomly assigned 1 : 1 to pembrolizumab plus FLOT or placebo plus FLOT. In this exploratory cohort, the primary endpoint was safety. Post hoc exploratory endpoints included pCR rate per the Mandard scale by blinded independent central review, EFS per RECIST v1.1 by investigator, and overall survival (OS). RESULTS: Overall, 203 participants were randomly assigned (pembrolizumab group, n = 100; placebo group, n = 103). Median time from randomization to data cut-off (16 February 2024) was 43.8 months (range, 36.5-69.8). Grade 3 or 4 treatment-related adverse event rates were 72.7% and 62.1%, respectively. Treatment-related deaths occurred in three participants (3.0%) in the pembrolizumab group and one participant (1.0%) in the placebo group. The pCR rate was 17.0% [95% confidence interval (CI), 10.2-25.8] in the pembrolizumab group and 6.8% (95% CI, 2.8-13.5) in the placebo group. Median EFS was not reached (NR; 95% CI, 28.2-NR) versus 31.6 months (95% CI, 22.8-NR), respectively [hazard ratio (HR) 0.75, 95% CI 0.50-1.13]. Median OS was NR in either group (HR 0.81, 95% CI 0.53-1.25); 36-month OS rates were 66.0% and 59.2%. CONCLUSION: Neoadjuvant/adjuvant pembrolizumab plus FLOT was feasible and showed an acceptable safety profile. Exploratory efficacy endpoints showed numerical trends consistent with those of the main cohort in locally advanced resectable gastric or gastroesophageal junction adenocarcinoma.