Abstract
We explored the rationale for treating glioblastoma (GBM) with regorafenib. In 103 newly diagnosed GBM patients, we assessed mutations, copy number variants (CNVs), fusions, and overexpression in 46 genes encoding protein kinases (PKs) potentially targeted by regorafenib or its metabolites and performed a functional enrichment analysis to assess their implications in angiogenesis. We analyzed regorafenib's binding inhibitory activity and target affinity for these 46 PKs and focused on a subset of 18 genes inhibited by regorafenib at clinically achievable concentrations and on 19 genes involved in angiogenesis. Putative oncogenic alterations were defined as oncogenic/likely oncogenic mutations, oncogenic fusions, CNVs > 5, and/or gene overexpression. Regorafenib did not target all 46 PKs. For the 46-gene set, 40 genes (86.9%) and 73 patients (70.8%) harbored at least one alteration in genes encoding targetable PKs, but putative oncogenic alterations were present in only 34 patients (33%). In the 18-gene set, 18 genes (100%) and 48 patients (46.6%) harbored alterations, but putative oncogenic alterations were detected in only 26 patients (25.2%). Thirty patients (29.1%) had oncogenic alterations in the 18-gene set and/or in angiogenesis-related genes. Around 33% of patients had oncogenic alterations in any of the 46 potential targets. Additionally, the suboptimal dosing of regorafenib, due to its poor penetration of the blood-brain barrier, may reduce the likelihood of effectively targeting certain PKs. Future use of multi-target drugs must be guided by a thorough understanding of target presence, effective inhibition, and the drug's ability to reach brain tumors at adequate concentrations.