Ramucirumab (RAM) plus FOLFOX as front-line therapy (Rx) for advanced gastric or esophageal adenocarcinoma (GE-AC): Randomized, double-blind, multicenter phase 2 trial.

Abstract

4004 Background: RAM, an anti-VEGFR2 monoclonal antibody, improved overall survival (OS) in 2 phase 3 trials in patients (pts) with previously treated gastric/gastroesophageal junction (GEJ) AC. We report the first assessment of RAM as 1st-line Rx for GE-AC. Methods: Pts with untreated metastatic or locally advanced unresectable GE-AC (PS ≤1) were randomized 1:1 to mFOLFOX6 plus RAM (8 mg/kg IV) v placebo (PL), q14d. Primary endpoint was progression-free survival (PFS), with 80% power to detect HR 0.71 (α =.3). Secondary endpoints included OS, response rate (RR), disease control rate (DCR), and safety. Results: 168 pts (RAM 84 v PL 84) enrolled at 47 US sites, 04/11 – 08/12. Pt characteristics: age (65 v 60); male (75% v 73%); gastric (23% v 24%), GEJ (31% v 27%), esophageal (46% v 49%); metastatic (95% v 94%). Median PFS 6.4 v 6.7 m (HR 0.98 [95% CI 0.69 – 1.37]; p =.89) and OS 11.7 v 11.5 m (HR 1.08 [0.73-1.58]). Subgroup analyses by primary tumor location: for esophageal, median PFS was 5.6 v 6.1 m (HR 1.30); for gastric/GEJ, PFS was 8.7 v 7.1 m (HR 0.77 [0.48 – 1.24]; p =.28) and OS 14.6 v 12.5 m. PFS rate at 3 m was higher in RAM v PL (89% v 75%, p =.020), but not at 6, 9, or 12 m. RR (CR, PR) 45% v 46%. DCR (SD, CR, PR) 85% v 67% (p =.008). Most common grade ≥3 adverse events (AEs): neutropenia (27% v 36%), fatigue (18% v 15%), neuropathy (9% v 11%). Grade ≥3 AEs of special interest were uncommon, except hypertension. Median cycles of OX were similar among arms (8.5 v 9.5), but cycles of 5FU or RAM/PL (both 9 v 11) were lower in RAM arm. Rx discontinuation for non-progressive disease (PD) was more common in RAM: pt/physician decision (27% v 10%), AEs (21% v 6%). In exploratory analyses that censored PFS at Rx discontinuation for non-PD, HR for PFS favored RAM arm (HR 0.76; p =.194), mainly in gastric/GEJ pts (PFS 9.3 v 7.6 m; HR 0.53 [0.29 – 0.97]; p =.036). Conclusions: Addition of RAM to FOLFOX did not improve median PFS but showed PFS difference at 3 m and improved DCR. Longer PFS in RAM v PL was observed in gastric/GEJ cancer pts. A higher non-PD discontinuation rate and lower drug exposure in RAM arm may have impacted PFS assessment. These data are critical for clinical development of RAM in gastric cancer. Clinical trial information: NCT01246960.