A phase II, randomized, non-comparative clinical trial of the effect of bevacizumab (BV) alone or in combination with irinotecan (CPT) on 6-month progression free survival (PFS6) in recurrent, treatment-refractory glioblastoma (GBM)

2010b Background: Effective 2nd-line therapies are absent in GBM leading to poor survival. VEGF upregulation is implicated in GBM tumorigenesis. In a single-arm PhII trial of BV + CPT in recurrent GBM, activity was demonstrated, warranting further investigation of clinical benefit of BV alone, and in combination with CPT (Vredenburgh J, JCO 2007). Methods: From June 2006 to Feb 2007, 167 patients (pts) with recurrent GBM were randomized 1:1 to BV (10 mg/kg q 2 wks) (Arm 1, n=85) or BV + CPT (340 mg/m2 if enzyme-inducing anti- epileptic drugs (AEDs) and 125 mg/m2 for non-AEDs) (Arm 2, n=82), stratified by KPS: 70–80, 90–100, and relapse (1st or 2nd). The co- primary endpoints were PFS6 (defined at 24-weeks) and objective response rate (ORR) determined by an independent radiology facility (IRF), with secondary endpoints of safety, PFS, and duration of response determined by IRF, and overall survival. The trial provided approximately 80% power to detect 13% improvement in PFS6 from an assumed rate of 15%, and 90% power to detect 13% improvement in ORR from assumed rates of 5% (Arm 1) and 10% (Arm 2) (2-sided significance level of 0.025). Results: Clinical outcomes are shown in Table 1. Median OS was based on 61.2% of deaths for BV alone and 67.1% of deaths for BV+CPT. One intracranial hemorrhage occurred in Arm 1 (Gr 1; 1.2%) and one in Arm 2 (Gr 4; 1.3%) during planned treatment period. The dose of corticosteroid required diminished from baseline to cycle 4 (Table 1). Benchmarking OS to that in a national GBM database is ongoing. Ä in KPS and neurocognitive function from baseline, and post-progression response will be presented. Conclusions: Both the PFS6 and OS of >9 months with BV alone or with CPT provides encouraging evidence of significant activity in this poor prognosis population. In addition, the reduction in corticosteroid use has the potential to provide benefit to patients. Randomized Subjects BV (n=85) BV + CPT (n=82) Median OS months (95% CI) 9.7 (8.2–11.8) 8.9 (7.8–11.9) PFS6 % (97.5% CI) 35.1 (23.2–47.0) 50.2 (36.6–63.8) ORR % (97.5% CI) 20 (12.7–29.5) 32.9 (23.4–43.5) Treated Subjects BV (n=84) BV + CPT (n=79) Grade >3 toxicities, % 47.6 67.1 Grade 5 AEs, % 2.4 1.3 BV discontinuation due to AEs, % 3.5 13.9 Corticosteroid dose (mg/kg) BV (n=84) BV + CPT (n=79) N, Mean (SD) Day 1 44, 7.3 (7.37) 46, 8.0 (4.96) Day 42 41, 5.0 (5.59) 44, 7.5 (5.71) Day 84 34, 3.1 (2.80) 34, 4.7 (3.68) Day 126 21, 2.7 (3.13) 21, 3.9 (2.60) Day 168 12, 1.1 (1.04) 15, 4.4 (4.13) No significant financial relationships to disclose.