Abstract Details

We hypothesized that proteomic profiling of plasma proteins collected in the acute setting from patients with stroke, TIA and stroke mimics is a feasible strategy to differentiate these diagnostic groups.

Transient ischemic attack (TIA) can often be difficult to distinguish from stroke mimics especially in the emergent setting. Blood based biomarkers to distinguish these entities could be immensely useful.

Blood was drawn acutely upon patient’s presentation in the ER and plasma samples were stored at -80’C until further processing. 29 age and sex matched samples (7 TIA, 10 ischemic stroke and 12 stroke mimics) were selected randomly. Top-14 most abundant (e.g. albumin, globulins) proteins were depleted. Pre- and post-depletion BCA and Coomassie gels were run to confirm adequate and equal depletion. Proteins were digested with lysC and trypsin, and processed for label-free quantitation mass spectrometry (LFQ-MS). Data were analyzed by differential expression and principal component analysis (PCA).

LFQ-MS of plasma samples identified 400 proteins of which 279 had <50% missing values and were included in the analysis. 51 proteins were differentially expressed across groups and PCA using these 51 proteins identified 3 PCs that explained 47.5% of the variance in the data (PC1:23.3 % PC2:14.9% PC3: 9.3%). PC1 differentiated TIA from mimics while PC2 differentiated stroke from TIAs and mimics. Higher levels of fibrinogen and complement activation differentiated TIA from mimics while higher levels of acute phase reactants (CRP, SAA1) differentiated TIA from ischemic stroke. 

We establish the feasibility of proteomic analysis of plasma as an approach to identify novel stroke and TIA biomarkers. We are undertaking comprehensive validation studies of these plasma biomarkers to develop diagnostic panels which could potentially serve to differentiate TIA from Stroke mimics in the emergency room setting.