Intravenous Glibenclamide Reduces Hypoxanthine, A Novel Predictor Of Brain Edema Formation: An Exploratory Analysis Of Games-Rp.

Introduction: To further characterize the effects of intravenous (IV) glibenclamide (glyburide; BIIB093) therapy on brain edema, we studied patients enrolled in the phase 2 GAMES-RP trial. We performed metabolomic analysis on serial blood samples to examine the relationship between treatment, plasma metabolites, and matrix metalloproteinase-9 (MMP-9) and midline shift (MLS). Methods: Metabolite profiling was performed on 399 serial plasma samples from the modified intention-to-treat GAMES-RP cohort (N=83). Blood samples were subjected to liquid chromatography-tandem mass spectrometry to measure 181 analytes. MMP-9 levels and MLS were measured as part of the original trial. Logistic regression was used to identify metabolites affected by IV glibenclamide treatment and repeated measures mixed effects models were used to examine association with MMP-9 and MLS. A decision tree-based machine learning algorithm, Random Forest (RF), was used to further verify leading candidates. Mediation analysis was performed to evaluate for metabolite mediators of IV glibenclamide’s effect on MMP-9 and MLS. Results: IV glibenclamide reduced plasma hypoxanthine level relative to treatment with placebo (24.7% reduction, p =0.003). Hypoxanthine was associated with increased MMP-9 level ( β =1.76, p =5.2x10 -18 ) and greater MLS ( β =0.081, p =2.8x10 -4 ). Multivariable analyses demonstrated these associations to be independent of age, gender, baseline NIHSS, DWI lesion volume, and tPA treatment. RF analysis confirmed hypoxanthine as a leading predictor of IV glibenclamide treatment, MMP9 and MLS. Hypoxanthine was a partial mediator of the effect of glibenclamide on MLS (22% mediation, Sobel test statistic -2.4, p =0.017) and a full mediator of MMP-9 (53% mediation, Sobel test statistic -2.6, p =0.009). Conclusion: Hypoxanthine was associated with increased MMP-9 and MLS, and was reduced by treatment with IV glibenclamide. As a biomarker implicated in free radical generation and blood-brain barrier breakdown after ischemia, hypoxanthine may be related to the effect of IV glibenclamide in preventing edema after ischemic stroke.