Event-free survival (EFS) in MATTERHORN: A randomized, phase 3 study of durvalumab plus 5-fluorouracil, leucovorin, oxaliplatin and docetaxel chemotherapy (FLOT) in resectable gastric/gastroesophageal junction cancer (GC/GEJC).

LBA5 Background: FLOT is a perioperative standard of care (SoC) in resectable GC/GEJC, yet recurrence rates remain high. Immune checkpoint inhibitors are approved in combination with chemotherapy in metastatic GC/GEJC, but not in the perioperative setting. The randomized, double-blind, global, Phase 3 MATTERHORN study (NCT04592913) assesses the combination of perioperative durvalumab (D) + FLOT vs placebo (P) + FLOT in participants (pts) with locally advanced, resectable GC/GEJC. The primary endpoint is EFS. Pathologic complete response (pCR) and overall survival (OS) are key secondary endpoints. The trial previously showed a statistically significant gain in pCR for D + FLOT. Here, we report efficacy and safety from the pre-planned interim analysis 2. Methods: Pts aged ≥18 years with histologically confirmed, resectable (Stage II–IVa per American Joint Committee on Cancer 8th edition) untreated G/GEJ adenocarcinoma were randomized 1:1 to D 1500 mg or P every 4 weeks (Q4W) on Day 1 + FLOT on Days 1 and 15 for 4 cycles (2 cycles each neoadjuvant/adjuvant), followed by D 1500 mg or P on Day 1 Q4W for 10 cycles. Pts were stratified by Asia vs non-Asia, clinical lymph node status (positive vs negative) and programmed cell death ligand-1 Tumor Area Positivity score (≥1% vs <1%). Data cutoff was Dec 20, 2024. EFS (time from randomization to progression, local or distant recurrence, or death) superiority for D + FLOT vs P + FLOT was assessed in all randomized pts by a stratified log-rank test (2-sided significance level threshold: 0.0239) on data according to RECIST v1.1 per BICR and/or locally by pathology testing. Results: In total, 948 pts were randomized to receive D + FLOT (n=474) or P + FLOT (n=474); median (m) follow-up duration was 31.5 months (mo). Demographic/baseline characteristics were generally similar across treatment arms. D + FLOT demonstrated a statistically significant improvement in EFS vs P + FLOT (hazard ratio [HR] 0.71; 95% confidence interval [CI], 0.58–0.86; p<0.001), mEFS was not reached (NR) with D + FLOT vs 32.8 mo with P + FLOT. The 24-mo EFS rate was higher for D + FLOT vs P + FLOT (Table). mOS was NR for D + FLOT vs 47.2 mo for P + FLOT (HR 0.78; 95% CI, 0.62–0.97; p=0.025; 33.9% maturity) and will be formally assessed at the final analysis. Maximum Grade 3 or 4 adverse event rates were similar between treatment arms; D + FLOT did not delay surgery or initiation of adjuvant therapy vs P + FLOT. Conclusion: D + FLOT demonstrated a statistically significant improvement in EFS vs P + FLOT in pts with resectable GC/GEJC, with an encouraging OS trend. These results support D + FLOT as a potential new global SoC for resectable GC/GEJC. Clinical trial information: NCT04592913 . D + FLOT (n=474) P + FLOT (n=474) mEFS (95% CI), mo NR (40.7–NR) 32.8 (27.9–NR) EFS rate (95% CI), %12 mo24 mo 78.2 (74.1–81.7)67.4 (62.9–71.6) 74.0 (69.7–77.8)58.5 (53.8–63.0)