Background: The addition of programmed cell death protein 1 (PD-1) inhibitors to chemotherapy (chemo) confers a survival advantage in 1L gastroesophageal cancers. However, long term outcomes remain poor. Dual PD-1 and anti-T-cell immunoglobulin and ITM domain (TIGIT) blockade increases tumor antigen-specific CD8+ T cell expansion, resulting in potent antitumor activity. The ongoing, multi-arm, global EDGE-Gastric trial (NCT05329766) is evaluating the safety and efficacy of various combinations of the anti-TIGIT Fc-silent monoclonal antibody (mAb) domvanalimab (D) and the anti-PD-1 mAb zimberelimab (Z) in patients (pts) with locally advanced unresectable or metastatic gastric (G)/gastroesophageal junction (GEJ)/esophageal (E) adenocarcinoma. Arm A1 is fully enrolled and results from this 1L arm are presented here. Methods: Patients with previously untreated G/GEJ/E adenocarcinoma received D 1600 mg intravenously (IV) every 4 weeks (Q4W) + Z 480 mg IV Q4W + FOLFOX (oxaliplatin 85 mg/m² IV, leucovorin 400 mg/m² IV, fluorouracil 400 mg/m² IV bolus + 2400 mg/m² continuous 46-48-hour IV infusion) Q2W. Primary endpoints are safety and objective response rate (ORR) per RECIST 1.1. Secondary endpoints include ORR by PD-L1 expression and PFS. Results: As of the data cutoff (5 June 2023), 41 pts were enrolled and had the opportunity to have ≥2 imaging assessments. There was an even distribution of pts accrued in Asia vs. rest of world and 63% of pts had gastric cancer. Median time on treatment was 23 weeks (1 to 40), 31 pts (76%) continue on protocol therapy, and there were no deaths on study. Intent-to-treat ORR was 59% (95% CI 42 to 74). Clinical outcomes by PD-L1 status are in the Table. The most common treatment-emergent adverse events (TEAE) were neutropenia (56% of pts), nausea (54%), and anemia (27%). Infusion related reactions were observed in 10% and none were deemed related to D/Z. Grade ≥3 TEAEs occurred in 66% pts, with 54% and 12% having grade ≥3 TEAEs related to FOLFOX and D/Z, respectively. Serious TEAEs occurred in 24% pts: 7% had serious TEAEs related to FOLFOX and no pts had serious TEAEs related to D/Z. TEAE related FOLFOX discontinuation occurred in 13 pts (32%), and in 1 pt due to D/Z. Conclusions: Addition of DZ to FOLFOX showed encouraging ORR and early PFS, particularly in pts with PD-L1-high tumors. The regimen is well tolerated, with a similar AE profile to anti-PD-1+FOLFOX. Updated safety and efficacy will be presented. The randomized phase 3 1L STAR-221 trial of DZ+chemo vs. SOC is underway. Clinical trial information: NCT05329766.

Tumor samples from 39 pts were available for central PD-L1 testing; TAP: Tumor Area Positivity.