Abstract 63: Plasma Biomarkers of Cavernous Angioma With Symptomatic Hemorrhage (CASH)

Introduction: Cerebral cavernous angiomas with a documented symptomatic hemorrhage (CASH) are associated with an exceptionally high risk of recurrent hemorrhage and serious morbidity. It is unclear if peripheral blood plasma biomarkers implicated in disease biology can differentiate CASH, and whether the same or different biomarkers can distinguish cases who would rebleed. Methods: Eighteen plasma molecules with postulated mechanistic roles in this disease were quantified in 114 patients and 12 healthy subjects. The best biomarker combination differentiating CASH in the prior year (+/- 30 days) was derived by minimizing the Akaike Information Criterion (AIC), and was validated using statistical and machine-learning simulations. This was compared to the biomarker predicting bleeding in the subsequent year (+/- 30 days), and both explanatory and predictive biomarkers were followed longitudinally in a subset of cases. Results: Nine plasma molecules had significantly different levels expressed in CASH patients (p<0·05, FDR corrected). The best biomarker (AIC=75·9) included a weighted combination of 4 of these (sCD-14, VEGF, CRP, and IL-10), and distinguished CASH patients with 76% sensitivity and 80% specificity (p=0·0003). A weighed combination of two of these compounds (sCD14 and VEGF) and two others (IL-1β and sROBO-4) predicted which patients would bleed in the subsequent year with 83% sensitivity and 93% specificity (p=0·001). The CASH biomarker increased in recovering patients (p=0·01), while the predictive biomarker decreased after a bleed (p=0·01). No significant change over time was observed in stable subjects in either biomarker. Conclusions: Our results suggest shared and unique biomarkers of recent symptomatic hemorrhage and of future bleeding in cavernous angiomas. This may be applied for risk stratification in clinical trials, and potentially developed as a tool in clinical practice.