Sponsored by
This retrospective study from Japan analyzed 50 patients with newly diagnosed, IDH-wildtype glioblastoma (GBM), of whom 18 received temozolomide (TMZ) alone and 32 received TMZ plus bevacizumab as part of initial therapy; most patients in the TMZ-only group later received bevacizumab at first recurrence. Most patients in the study also received carmustine (aka BCNU or Gliadel) wafer implantation at surgery. Overall, adding bevacizumab upfront improved progression-free survival (PFS) but not overall survival, consistent with prior trials. When patients were stratified by tumor COX-2 expression, those with high COX-2 expression experienced significantly longer PFS and overall survival when bevacizumab was given upfront compared with TMZ alone (median PFS 22 v 8 months; median OS 25 vs 18 months), while no benefit was observed in patients with low COX-2 expression (median PFS 12 vs 15 months; median OS 24 vs 26 months). These results are biologically plausible and suggest that COX-2 may help identify a subset of newly diagnosed patients who benefit from early bevacizumab; however, given the retrospective design, small sample size, treatment crossover, use of BCNU wafers, and potential subjectivity of the immunohistochemistry biomarker scoring, these findings should be followed up with prospective validation. Also, the results may not apply to recurrent disease, where prior therapies can substantially alter tumor angiogenic biology.