4003 Background: DANTE evaluates atezolizumab in the perioperative treatment of resectable gastric or gastroesophageal junction (GEJ) adenocarcinoma in combination with FLOT. Here, we report interim results. Methods: DANTE is a multicenter, investigator-initiated, phase IIb trial. Patients (pts) with resectable adenocarcinoma of the stomach and GEJ (≥cT2 and/or N+) were randomized to receive 4+4 cycles of periop. FLOT chemotherapy (arm B) or the same regime with additional atezolizumab at 840 mg, q2w, followed by atezolizumab monotherapy for 8 cycles at 1200 mg, q3w (arm A). The primary endpoint is progression-free survival. The secondary endpoints surgical outcome (pTNM, R0 resection rate and periop. morbidity/mortality), path. regression and safety are reported here. TNM stage was evaluated by local pathologists and path. regression (Becker-Classification) by local and central pathologists. PD-L1 and MSI status were centrally evaluated. Results: In total, 295 pts were randomized (A, 146; B, 149) with baseline characteristics as follows: median age 61y, male 74%, intestinal type 42%, GEJ 61%, cT3/4 77%, N+ 78%. Twenty-five pts (8.5%) were MSI; 50% had PD-L1 CPS ≥1, 23% PD-L1 CPS ≥5 and 15% PD-L1 CPS ≥10. Pre-op FLOT cycles were completed in 93% of pts and post-op cycles in 43% of pts, with no difference between arms. Surgical morbidity (A, 45%; B, 43%) and mortality (overall 2.5%) were comparable between arms, as were R0-resection rates (arm A, 92% vs. arm B, 91%). Downsizing favored arm A vs B (pT0, 23% vs 15%; pN0, 68% vs 54%). Increases in path. regression rates were seen, particularly with higher PD-L1 expression (Table). Conclusions: The analysis shows beneficial effects of atezolizumab combined with FLOT vs FLOT alone on path. stage and path. regression that seem to be more pronounced with higher PD-L1 expression. Sponsor: Institut für Klinische Krebsforschung IKF am Krankenhaus Nordwest. EudraCT: 2017-001979-23. Clinical trial information: NCT03421288. [Table: see text]
