Currently, the standard treatment for newly diagnosed glioblastoma multiforme (GBM) is maximal safe surgical resection followed by radiation therapy with concurrent and adjuvant temozolomide. However, disease recurs in almost all patients, and the optimal salvage treatment for recurrent GBM remains unclear. We conducted a systematic review and meta-analysis of published clinical trials to assess the efficacy and toxicities of angiogenesis inhibitors alone as salvage treatment in these patients.
Trials published between 1994 and 2015 were identified by an electronic search of public databases (MEDLINE, EMBASE, Cochrane library). Demographic data, treatment regimens, objective response rate (ORR), median progression-free survival (PFS), median overall survival (OS), 6-months PFS rate, 1-year OS and grade 3/4 toxicities were extracted. We also compared the main outcomes of interest between bevacizumab and other angiogenesis inhibitors. All analyses were performed using Comprehensive Meta Analysis software (Version 2.0).
A total of 842 patients were included for analysis: 343 patients were treated with bevacizumab, 386 with other angiogenesis inhibitors and 81 with thalidomide. The pooled ORR, 6-months PFS, and 1-year OS for recurrent GBM patients receiving angiogenesis inhibitors was 20.1%, 19.5% and 29.3%, respectively. The use of single agent bevacizumab in recurrent GBM significantly improved ORR and 6-months PFS when compared to other angiogenesis inhibitors [relative risk (RR) 2.93, 95% CI 1.38–6.21; p = 0.025; and RR 2.36 95% CI 1.46–3.82; p<0.001, respectively], while no significant difference in 1-year OS was found between the two groups ( p = 0.07). when compared to thalidomide, bevacizumab treatment in recurrent GBM significantly improved ORR (RR 6.8, 95%CI: 2.64–17.6, p<0.001), but not for 6-months PFS ( p = 0.07) and 1-year OS ( p = 0.31). As for grade 3/4 toxicities, the common toxicity was hypertension with pooled incidence of 12.1%, while high-grade thromboembolic events (2.2%), hemorrhage (5.1%) and GI perforation (2.8%) associated with angiogenesis inhibitors were relatively low.