Background: Amivantamab (Ami) is a bispecific antibody targeting epidermal growth factor receptor (EGFR) and mesenchymal epithelial transition receptor (Met) and has immune cell directing activity. Gastric cancer (GC), gastroesophageal junction (GEJ) cancer and esophageal cancer (EC) have been known to express EGFR and Met and these expressions have correlated with poor prognosis. Methods: This open-label, multicenter phase 2 study aims to evaluate the antitumor activity and safety of Ami monotherapy in patients (pts) with advanced or unresectable GC, GEJ cancer and EC who had received at least 2 prior lines (GC/GEJ) or 1 prior line (EC) of systemic therapy. Pts who express varying degrees of EGFR, Met, or both as determined by immunohistochemistry (IHC) were enrolled. Ami was administered intravenously at the recommended Phase 2 dose (RP2D) of 1,050 mg (or 1,400 mg if body weight ≥80 kg) in 28-day cycles. The primary endpoint was objective response rate (ORR) (RECIST 1.1). No hypothesis was planned but approximately 30 pts were to be enrolled in GC and EC cohorts, respectively. Secondary endpoints included disease control rate (DCR), pharmacokinetics (PK) and safety. Results: At the data cut-off of Jul-2023, a total of 59 pts received the RP2D. The 29 and 30 pts had GC/GEJ (all adenocarcinoma) and EC (29 squamous-cell and 1 adenocarcinoma), respectively. Median ages were 66 in GC/GEJ and 60 in EC, and 82.8% and 80.0% of pts were male, and median durations of study treatment were 0.99 and 2.64 months, respectively. The ORRs were 4.3% for GC/GEJ and 10.7% for EC and DCRs were 26.1% and 67.9%, respectively. The Met-high-expression (IHC score ≥2+) tended to be associated with better response of ORR 13.3% and DCR 80.0% compared to Met-low-expression (IHC score 1+/0) of ORR 7.7% and DCR 53.8% in EC. Transcriptomic analysis results from paired baseline/on-treatment EC biopsies will be presented. All 59 pts had adverse events (AEs) and 64.4% of them had AEs with the maximum grade of ≤2. Serious AEs occurred in 27.1% of pts. Conclusions: Ami showed clinical meaningful antitumor activity in EC, especially in Met-high-expression tumors. There was minimal monotherapy activity in GC. No new safety signals were identified in pts with GC and EC, which appeared to be acceptable. Clinical trial information: NCT04945733.