Rilotumumab in combination with epirubicin, cisplatin, and capecitabine as first-line treatment for gastric or oesophagogastric junction adenocarcinoma: an open-label, dose de-escalation phase 1b study and a double-blind, randomised phase 2 study

Background Dysregulation of the hepatocyte growth factor (HGF)/MET pathway promotes tumour growth and metastasis. Rilotumumab is a fully human, monoclonal antibody that neutralises HGF. We aimed to assess the safety, efficacy, biomarkers, and pharmacokinetics of rilotumumab combined with epirubicin, cisplatin, and capecitabine (ECX) in patients with advanced gastric or oesophagogastric junction cancer. Methods We recruited patients (>= 18 years old) with unresectable locally advanced or metastatic gastric or oesophagogastric junction adenocarcinoma, an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1, who had not received previous systemic therapy, from 43 sites worldwide. Phase 1b was an open-label, dose de-escalation study to identify a safe dose of rilotumumab (initial dose 15 mg/kg intravenously on day 1) plus ECX (epirubicin 50 mg/m(2) intravenously on day 1, cisplatin 60 mg/m(2) intravenously on day 1, capecitabine 625 mg/m(2) twice a day orally on days 1-21, respectively), administered every 3 weeks. The phase 1b primary endpoint was the incidence of dose-limiting toxicities in all phase 1b patients who received at least one dose of rilotumumab and completed the dose-limiting toxicity assessment window (first cycle of therapy). Phase 2 was a double-blind study that randomly assigned patients (1:1:1) using an interactive voice response system to receive rilotumumab 15 mg/kg, rilotumumab 7.5 mg/kg, or placebo, plus ECX (doses as above), stratified by ECOG performance status and disease extent. The phase 2 primary endpoint was progression-free survival (PFS), analysed by intention to treat. The study is registered with ClinicalTrials.gov, number NCT00719550. Findings Seven of the nine patients enrolled in the phase 1b study received at least one dose of rilotumumab 15 mg/kg, only two of whom had three dose-limiting toxicities: palmar-plantar erythrodysesthesia, cerebral ischaemia, and deep-vein thrombosis. In phase 2, 121 patients were randomly assigned (40 to rilotumumab 15 mg/kg; 42 to rilotumumab 7.5 mg/kg; 39 to placebo). Median PFS was 5.1 months (95% CI 2.9-7.0) in the rilotumumab 15 mg/kg group, 6.8 months (4.5-7.5) in the rilotumumab 7.5 mg/kg group, 5.7 months (4.5-7.0) in both rilotumumab groups combined, and 4.2 months (2.9-4.9) in the placebo group. The hazard ratio for PFS events compared with placebo was 0.69 (80% CI 0.49-0.97; p=0.164) for rilotumumab 15 mg/kg, 0.53 (80% CI 0.38-0.73; p=0.009) for rilotumumab 7.5 mg/kg, and 0.60 (80% CI 0.45-0.79; p=0.016) for combined rilotumumab. Any grade adverse events more common in the combined rilotumumab group than in the placebo group included haematological adverse events (neutropenia in 44 [54%] of 81 patients vs 13 [33%] of 39 patients; anaemia in 32 [40%] vs 11 [28%]; and thrombocytopenia in nine [11%] vs none), peripheral oedema (22 [27%] vs three [8%]), and venous thromboembolism (16 [20%] vs five [13%]). Grade 3-4 adverse events more common with rilotumumab included neutropenia (36 [44%] vs 11 [28%]) and venous thromboembolism (16 [20%] vs four [10%]). Serious adverse events were balanced between groups except for anaemia, which occurred more frequently in the combined rilotumumab group (ten [12%] vs none). Interpretation Rilotumumab plus ECX had no unexpected safety signals and showed greater activity than placebo plus ECX. A phase 3 study of the combination in MET-positive gastric and oesophagogastric junction cancer is in progress.