Bevacizumab (BV) continuation following BV progression: Meta-analysis of five consecutive recurrent glioblastoma (GBM) trials.

Abstract

2030 Background: Although BV benefits most recurrent GBM patients, the duration of benefit is limited and subsequent survival after BV progression is poor due to lack of effective therapeutic options. Methods: We performed a meta-analysis of five consecutive phase II BV salvage trials conducted at our institution to evaluate potential prognostic factors and to assess outcome following BV trial progression for recurrent GBM patients. Results: The five BV salvage trials included similar eligibility, treatment and assessment criteria and enrolled 172 recurrent GBM patients. Outcome across the trials was comparable and traditional prognostic factors including age, KPS and degree of prior treatment did not predict outcome. Following progression on BV trial therapy, 95 patients (68%) received additional therapy including 42 patients who were initially treated with a non-BV regimen and 53 who received a BV regimen. Patients who received non-BV therapy did not differ from those who received BV therapy with regard to age, number of prior episodes of progression, time from original diagnosis or duration on BV trial therapy. The median overall survival (OS) and OS at 6 months (OS-6) for patients who received a BV regimen after progression on a BV trial was 6.1 months (95% CI: 5.2, 7.6) and 51.1% (95% CI: 36.7, 63.8), while the median OS and OS-6 for patients treated with a non-BV regimen were 4.5 months (95% CI: 2.5, 5.4) and 31.0% (95% CI: 17.8, 45.0), respectively (p=0.0135). After adjusting for known risk factors, BV therapy was the only factor associated with OS for patients undergoing further therapy after BV trial progression (hazard ratio 0.564; p-value 0.0287). Conclusions: Among recurrent GBM patients, clinical factors may poorly predict outcome for BV therapy. Outcome following BV failure is poor, but continuation of BV improves OS compared to available non-BV therapy.