Randomized, double-blind, placebo-controlled phase II study of AMG 386 in combination with cisplatin and capecitabine (CX) in patients (pts) with metastatic gastroesophageal adenocarcinoma.

Abstract

66 Background: AMG 386, a first-in-class investigational peptide-Fc fusion protein (peptibody), blocks angiogenesis via inhibiting the interaction between angiopoietins-1 and -2 and the Tie2 receptor. We evaluated the efficacy and tolerability of AMG 386 or placebo plus CX in the first-line treatment of metastatic gastroesophageal adenocarcinoma. Methods: Pts with confirmed metastatic adenocarcinoma of the stomach, gastroesophageal junction or distal esophagus were randomized 1:1:1 to receive CX (cisplatin, 80 mg/m2 IV Q3W; capecitabine, 1,000 mg/m2 orally BID for 14 days Q3W) plus AMG 386 10 mg/kg (Arm A), 3 mg/kg (Arm B), or placebo (Arm C) IV QW. The primary endpoint was progression-free survival (PFS). Secondary endpoints included objective response rate (ORR; in pts with measurable disease), adverse events (AEs), and pharmacokinetics (PK). Results: 171 pts were randomized (Arm A/B/C, n = 56/59/56). Efficacy results are summarized in the table. The incidence of grade ≥ 3 AEs in Arms A/B/C was 80/84/75%. Serious AEs occurred in 73/60/47% and serious AEs grade ≥ 3 in 66/60/43% of pts. AEs in Arms A/B/C included abdominal pain (30/40/17%; grade ≥ 3, 18/3/4%), peripheral edema (13/29/6%; grade ≥ 3, 0/2/0%), venous thromboembolic events (20/22/19%; grade ≥ 3, 20/19/17%), and pulmonary embolism (9/3/15%; grade ≥ 3, 9/2/13%). Median AMG 386 Cmax and Cmin values at steady state after CX coadministration were dose-proportional. Coadministration with CX did not markedly affect AMG 386 exposure. Conclusions: In this study, AMG 386 plus CX did not significantly improve PFS or ORR over placebo plus CX in this patient population. The toxicity of the combination of AMG 386 plus CX, compared with placebo, was greater but manageable. No unexpected AEs occurred. [Table: see text] [Table: see text]