Phase Ib study of olinvacimab (O) with pembrolizumab (P) in patients with recurrent glioblastoma (rGBM).

Abstract

e14545 Background: Recurrent GBM is difficult to treat. Single agent checkpoint blockade has not improved outcomes. Angiogenesis is a rational drug target for rGBM and targeting angiogenesis may benefit pseudoprogression and cerebral oedema. O is a fully human monoclonal antibody (MAB) which binds to Vascular Endothelial Growth Factor Receptor 2 (VEGFR2) with antiangiogenic and antitumour effects. P is an anti-PD1 MAB. This study aimed to identify the safety and tolerability of O combined with P and to establish a recommended phase 2 combination dose. Methods: From January to October 2019, we conducted a two-site, single arm, open-label study of O with P in patients with rGBM. Eligible patients (pts) were ≥18 years with at least one RANO-measurable lesion, KPS≥80, and had completed standard chemoradiotherapy and had no contraindications to O or P. No prior bevacizumab was allowed. A modified Toxicity Probability Interval design was used. Pts received O 12 mg/kg day 1/8/15 q21d (dose level 1) or O 16mg/kg day 1/8/15 q21d (dose level 2) in combination with P 200mg flat dose day 1 q21d. Pts were reviewed weekly and underwent DCE-MRI at baseline and 6-weekly. Treatment continued to progression, toxicity or withdrawal. Results: 9 pts, median age 53 (range 34-67) were recruited and received at least one study treatment. Median time since diagnosis was 15.6 months. 7 (78%) had KPS 90 and 2 (22%) KPS 80. 3 pts received O 12mg/kg with P, completing a median 3 cycles (range 2-6). As no Dose Limiting Toxicities (DLTs) were seen, 3 pts were treated with O 16mg/kg with expansion total 6 when no DLTs were observed. At 16mg/kg, a median of 2 treatment cycles was received (range 2-6). Treatment was ceased due to progressive disease (PD) in 8 pts with one ongoing at data cutoff. No DLTs were observed in any pts. Three grade 3 treatment emergent adverse events (TEAEs) were noted (blurred vision, fatigue, and seizure). Hemangioma is a known toxicity of O and was seen in 6 pts, with 11 grade 1 and 1 grade 2 event. 4 pts (44%) had stable disease (SD) and 5 pts (56%) had PD as best response. Conclusions: The combination of O with P was safe and tolerable at the full single agent dose of each drug. No DLTs were observed. The combination did not show efficacy in this setting. Clinical trial information: NCT03722342.