Glycine antagonist (GV150526) in acute stroke: a multicentre, double-blind placebo-controlled phase II trial

GV150526 is a novel glycine antagonist at the NMDA receptor complex. It is a potent neuroprotective agent in animal models of acute stroke including permanent middle cerebral artery occlusion in the rat. GV150526 was very well tolerated in early human studies. The purpose of this randomised, double-blind, multicentre, placebo-controlled trial was to assess the safety and population pharmacokinetics of GV150526 in patients with a clinical diagnosis of acute stroke. Exploratory assessment of efficacy, quality of life and resource utilisation was also undertaken. Upon clinical diagnosis of acute stroke within 12 h of onset of symptoms, patients were treated with a loading dose of 800 mg GV150526 (or placebo), followed by 5 maintenance doses of 400 mg GV150526 (or placebo) given every 12 h over 3 days. Following observation of asymptomatic hyperbilirubinaemia, the maintenance dose was reduced mid-study to 200 mg. CT/MRI scanning was not mandatory prior to treatment. The study treated 128 patients (38 with GV 800 mg/400 mg, 48 with GV 800 mg/200 mg and 42 with placebo). Fewer patients with mild stroke (NIH scores < or =5) were enrolled in the GV150526-treated groups than in the placebo group (placebo 38%, GV 800 mg/400 mg 18%, GV 800 mg/200 mg 15%). There was also an imbalance in the proportion of patients with haemorrhagic strokes (placebo 5%, GV 800 mg/400 mg 3%, GV 800 mg/200 mg 15%). Mortality at 1 month was unbalanced between treatment groups, being 10, 18 and 17% in the placebo, GV 800 mg/400 mg and GV 800 mg/200 mg groups, respectively (no significant difference). Similarly, adverse events, though consistent with an acute stroke population, appeared more often in the GV 800 mg/200 mg group. Functional outcomes at 1 month also showed imbalances, the percentage of patients with a Barthel Index score of > or =95 at 1 month being 52, 39 and 27% in the placebo, GV 800 mg/400 mg and GV 800 mg/200 mg groups, respectively. These results probably reflect a prognostically significant baseline difference between the groups rather than the effect of GV150526. GV150526 was generally well tolerated in patients with a clinical diagnosis of acute stroke and formal efficacy studies were considered justified.