A multicenter randomized phase III study for recurrent glioblastoma comparing bevacizumab alone with dose-dense temozolomide followed by bevacizumab: JCOG1308C.

2046 Background: Temozolomide (TMZ) is an alkylating agent commonly used as the standard therapy for newly diagnosed glioblastoma (GBM), with the DNA repair enzyme O 6 -methylguanine-DNA methyltransferase (MGMT) serving as a key prognostic and predictive factor. Despite treatment, GBM almost always recurs with limited therapeutic options, leading to poor prognosis. Since MGMT is consumed during the repair of TMZ-induced O 6 -methylguanine lesions in DNA, dose-intensified TMZ regimens are designed to deplete MGMT, thereby enhancing tumor sensitivity to TMZ. Bevacizumab (BEV), an anti-VEGF agent, has shown efficacy in recurrent GBM (re-GBM), but no effective therapies exist after BEV failure. Introducing an active agent before BEV may improve outcomes. To test this, we conducted a multicenter, phase III study comparing BEV monotherapy with dose-dense TMZ (ddTMZ) followed by BEV in re-GBM. Methods: Patients (pts) aged 20-75 years with KPS ≥60 and histologically confirmed GBM at first recurrence were enrolled from 31 Japanese hospitals. Participants were randomized to BEV monotherapy (10 mg/kg every 2 weeks; arm A) or ddTMZ (120-150 mg/m 2 , 7 days on/7 days off) followed by BEV at progression (arm B). Treatment continued until progression or unacceptable toxicity. The primary endpoint was overall survival (OS). A planned sample size of 146 pts provided 70% power to detect a hazard ratio (HR) of 0.73 (median OS (mOS): 8 vs. 11 months) at a one-sided alpha of 10%. MGMT promoter methylation and IDH mutation status were analyzed. Results: From July 2016 to April 2022, 146 pts (73 per arm) were randomized. MGMT promoter methylation was observed in 78 pts, while 49 were unmethylated. IDH1 mutations were identified in 8 of 129 pts to be tested. The mOS was 11.0 months (95% CI: 9.0-12.8) in arm A and 10.8 months (95% CI: 8.6-12.5) in arm B, with no significant difference (HR 0.922, 95% CI: 0.655-1.297, one-sided p = 0.320). No significant OS difference was observed between arms based on MGMT methylation status. The median progression-free survival (PFS) was 4.0 months (95% CI: 3.8-5.7) in arm A and 2.0 months (95% CI: 1.9-2.1) in arm B (HR 1.632, 95% CI: 1.168-2.281). Most pts in arm B exhibited progression at their first MRI. From the start of BEV treatment, mOS was 10.8 months (95% CI: 8.8-12.6) in arm A, and 8.0 months (95% CI: 6.1-9.1) in arm B. Grade 3-4 adverse events included hypertension (19.4%) in arm A and lymphopenia (52.1%) and leukopenia (8.2%) during ddTMZ in arm B. Grade 4 toxicities were rare. Conclusions: While ddTMZ was well-tolerated, this study failed to demonstrate a survival benefit for ddTMZ followed by BEV in re-GBM. BEV remains the preferred treatment at first recurrence. Further research is needed to develop effective therapies beyond the current standard for re-GBM. Clinical trial information: NCT02761070 .