Clinical development and evaluation of a VEGF-D assay in plasma from patients with metastatic colorectal cancer in the RAISE study

Background Vascular endothelial growth factor (VEGF)-D was identified as a potential predictive biomarker for ramucirumab efficacy in second-line metastatic colorectal cancer using a research use only (RUO) assay. We describe results with a new assay for detecting VEGF-D in human plasma. Methods In RAISE (Clinical Trial Registration: NCT01183780), 1072 patients were randomized 1:1 to ramucirumab or placebo plus FOLFIRI. All patients were then randomized 1:2 to marker exploratory (ME) and marker confirmatory (MC) groups, and those with plasma samples were analyzed accordingly. A new assay validated for investigational use only (IUO) was used to measure VEGF-D levels in plasma, which were analyzed for correlation with overall and progression-free survival (OS/PFS). IUO assay data were compared with historical RUO assay data. Results ME subset analyses determined the optimal cutpoint of 5.4 ng/mL for defining high/low VEGF-D subgroups. In the combined ME/MC placebo arms, OS/PFS were numerically greater for patients with low vs high VEGF-D (OS: 12.8 vs 11.1 months; PFS: 5.6 vs 4.2 months). In patients with high VEGF-D, ramucirumab vs placebo demonstrated a numerically greater improvement in OS and PFS. Differential efficacy by VEGF-D level was statistically significant for PFS, but not OS. Conclusion In patients with high VEGF-D, ramucirumab demonstrated a greater improvement in OS and PFS vs placebo; however, baseline VEGF-D level was not predictive of ramucirumab OS benefit using VEGF-D assay for IUO. The RAISE intent-to-treat results remain valid.