838TiP - Randomized phase II trial of weekly paclitaxel + ramucirumab versus weekly nab-paclitaxel + ramucirumab for unresectable advanced or recurrent gastric cancer with peritoneal dissemination refractory to first-line therapy: WJOG10617G/P-SELECT

Ramucirumab (RAM) with weekly paclitaxel (wPTX) is the recommended standard second-line therapy for advanced and recurrent gastric cancer (GC). In a recent phase II study, wNab-PTX + RAM combination for GC showed promising safety and efficacy, but it remains unclear whether wPTX + RAM or wNab-PTX + RAM should be used as second-line therapy. The previous ABSOLUTE Trial compared triweekly Nab-PTX, wNab-PTX, and wPTX in patients with advanced/recurrent GC as second-line therapy. Subgroup analysis of subjects with peritoneal dissemination (52.8%) showed that the hazard ratios (HRs) for overall survival (OS) and progression-free survival (PFS) in the wNab-PTX arm compared with the wPTX arm were 0.78 and 0.64, respectively. Therefore, we are investigating whether wNab-PTX + RAM is more effective than wPTX + RAM for patients with peritoneal dissemination. WJOG10617G/P-SELECT is an open-label, randomized, phase II study evaluating wPTX + RAM (arm A) versus wNab-PTX + RAM (arm B). Key eligibility criteria: 1) Histological diagnosis of primary GC, 2) unresectable or recurrent GC, 3) patients with peritoneal dissemination, 4) intolerant or refractory to first-line therapy including a fluoropyrimidine, and 5) ECOG Performance Status (PS) 0-2. Stratification factors are: institution, PS, and severity of ascites. Arm A comprises PTX 80mg/m2(day 1,8,15) and RAM 8mg/kg (day 1,15) every 4 weeks. Arm B comprises Nab-PTX 100mg/m2(day 1,8,15) instead of PTX. The primary endpoint is OS, and the main secondary endpoints are PFS, objective response rate (ORR), safety, and neuropathy-specific quality of life (QOL). SPARC and Caveolin-1 (Cav-1) expression in tumor tissues will be evaluated as predictive biomarkers for efficacy. Pharmacokinetics of RAM will also be assessed. The required sample size for the full analysis was 50 patients per group to maintain a probability of 70% or higher that the HR for OS in arm B would be lower than 0.90. Enrollment was therefore set at 105 subjects. This study has enrolled 26 patients as of March 31, 2019. jRCTs031180022, 27 September 2018. WJOG (West Japan Oncology Group). Taiho Pharmaceutical Co., Ltd. All authors have declared no conflicts of interest.