Exposure and clearance of bevacizumab in patients with first-line advanced gastric cancer.

Abstract

2603 Background: Bevacizumab (BEV), rhuMAB VEGF, in combination with chemotherapy (CT) is currently approved in various types of cancer. Recently, BEV+CT was evaluated in 1st line advanced gastric cancer (AGC) in a double-blind randomized phase 3 trial (AVAGAST), but failed to meet the primary OS endpoint (HR=0.87; p=0.1002). BEV pharmacokinetics (PK) is well established by a reference population PK (PPK) model, and BEV PK is consistent across multiple cancer indications. However, BEV PK in AGC was never evaluated. We aimed to assess BEV exposure (EXP) and PK in AGC and to explore the influence of demographic, prognostic and biochemical (DPB) factors. Methods: BEV concentrations (Cp) were measured from plasma samples collected following disease progression from 182 patients (7.5mg/kg Q2W) in AVAGAST. Expected BEV Cp [median and 90% prediction interval (PI)] were simulated using the PPK model and compared to the observed BEV EXP. BEV clearance (CL) in AGC was estimated using NONMEM and compared between subgroups stratified by DPB factors. Results: All DPB factors of AGC are similar to those in other cancers except for lower body weight. No cachexia was observed. BEV Cp was detectable in 162 patients. 85% of observed BEV Cp was below the median BEV Cp simulated using PPK model and 38% below the lower limit of the 90% PI. Median BEV CL in AGC was 4.5 L/day/kg vs. 3 L/day/kg in other cancers. BEV CL was significantly higher (p=0.0012) in patients without prior gastrectomy (4.9 L/day/kg, n=120) than those with gastrectomy (4.1 L/day/kg, n=42). CL appeared to be higher in AGC patients with higher ECOG scores, lower albumin, more metastatic sites, poorer response and later stage AGC. Tumor location (GE Jct vs. Stomach), VEGF-A and ethnicity (Asian vs. Non-Asian) did not correlate with BEV CL. Conclusions: Overall, AGC patients exhibited significantly lower BEV EXP due to a 50% increase in BEV CL vs. other cancers. In addition, BEV is cleared significantly faster in patients without prior gastrectomy. The low BEV EXP in AGC could potentially reduce the effect of BEV in AGC. The underlying mechanism for faster BEV CL in AGC is unknown and warrants further research and potential BEV dose modifications.