P08.09 Axitinib for the treatment of patients with recurrent glioblastoma, final results from a randomized phase II clinical trial

Abstract

BACKGROUND: vascular endothelial growth factor receptor (VEGFR) signal transduction mediates glioblastoma (GB) associated neo-angiogenesis. The AXIG trial (NCT01562197) is a randomized clinical trial that initially investigated the activity of axitinib as a monotherapy (AXI-arm) versus physicians best alternative choice of therapy (Duerinck et al. JNO Mar 2016) and following amendment versus axitinib plus lomustine (Duerinck et al. ASCO AM 2016). METHODS: an updated pooled analysis was made of the clinical outcome of all patients who initiated treatment with axitinib mono-therapy or axitinib in combination with lomustine in the AXIG trial. RESULTS: between August 2011 and July 2015, a total of 78 pts were enrolled in the trial and initiated treatment with axitinib monotherapy (N: 50; AXI) or axitinib plus lomustine (N: 28; AXILOM). Median age was 55y [range 18-80], 50M/28F, 19, 29, 19, 7 and 4 pts had a WHO-PS of 0, 1, 2, 3 and 4 respectively. Baseline characteristics were well balanced between study arms (AXI vs AXILOM). All pts had failed prior surgery, RT and TMZ. Thirteen pts in the AXI-arm crossed-over at the time of progression (AXIseqLOM). Treatment was generally well tolerated. AXILOM pts were at higher risk for grade 3/4 adverse events (most frequent gr3/4 AE on AXILOM vs AXI were: thrombocytopenia 3-vs 0 pts, hypertension 2 vs 2 pts, anorexia 3 vs 3 pts). The best overall tumor response (by RANO criteria) was 2 CR/11 PR/11 SD/26 PD in the AXI arm (BORR 26%) vs. 1 CR/9 PR/3 SD/13 PD in the AXILOM-arm (BORR 38%). Adding CCNU to axitinib after progression (AXIseqLOM) resulted in 1 pt experiencing PR and 8pts experiencing SD. Six-mths PFS and OS rates were respectively 26% (95% CI 14-38) vs 17% (95% CI 2-32), and 54% (95% CI 40-68) vs. 53% (95% CI 34-73) for pts treated in the AXI vs. AXILOM-arms; median PFS and OS were respectively 12 vs. 16 wks and 25 vs. 27 wks. When the time-to-second progression was taken into account for the 9 AXIseqLOM patients who experienced PR or SD following the addition of lomustine at first PD on axitinib, 6-mths PFS was 39% (95% CI 21-57). No significant correlation was found between clinical outcome measures and the MGMT promoter methylation status or IDH1/2 mutation status. CONCLUSION: Axitinib has single-agent activity in patients with recurrent GB; upfront combination of axitinib and LOM did not significantly increase progression-free or overall survival in patients with recurrent glioblastoma. Adding CCNU to axitinib after progression seems to result in improved PFS compared to upfront combination therapy.