Diagnostic and predictive value of circulating microRNAs and their regulation of BDNF signaling pathway in stroke patients

Abstract

<b>Abstract ID 13936</b> <b>Poster Board 456</b> <b>Aim:</b> We aimed to identify and validate circulating miRNAs, modulated in ischemic stroke (IS) patients in order to distinguish the most specific as biomarkers to facilitate diagnosis and prognosis. <b>Method:</b> Two microarray miRNA profilings were done and validated by RT-qPCR to explore (i) diagnostic miRNAs (29 acute IS patients vs 30 controls), and (ii) predictive miRNAs (24 patients with single stroke vs 24 patients with multiple stroke or TIA-NAVIGATE-ESUS cohort (ClinicalTrials.gov, NCT02313909). Stat: TAC software, R using Signal information, FDR correction, logistic reg, Mann-Whitney, t-test and calculated AUC using ROCp R. <b>Results:</b> Microarray data identified 404 differentially expressed (DE) probes. 146 up- and 258 downregulated. Target prediction of those miRNAs showed 67 up- and 125 downregulated miRNAs mapped by multiMir R package (fig1a). Targets of upregulated top miRNAs were most associated with BDNF, IL-2 signaling pathway, FSH regulation of apoptosis, Axon guidance and TGF-beta regulation of EC matrix. Downregulated miRNAs were most associated with Axon guidance, Neuronal system and Signaling by NGF. Targets of up/down regulated top miRNAs most associated with CVD-related terms are presented in fig1b. ANKRD52, AGO1 were targeted by all types of DE miRNAs. Most susceptible to regulation by up-regulated miRNAs: ANKRD12 and HIF1A and down-regulated miRNAs: GNAI2 and GRIN1 (fig1c). <b>1. Diagnostic analysis:</b> miR-18a-5p was higher in IS patients both at day1 and day7 compared to control (p=0.001, p=0.009, res). MiR-199a-5p was higher in acute IS, and stayed upregulated at day7 (p&lt;0.001, p=0.002, res). MiR-4467 was lower in the IS patients at day1 compared to control (p&lt;0.001). MiR-3135b was downregulated in IS patients at day1 and day7 compared to control (p&lt;0.001,p&lt;0.001, res). ROC curve showed diagnostic value for all studied miRNAs for acute stage of IS (p&lt;0.01), see fig1d). <b>2. Predictive analysis:</b> MiR-4786 (AUC=0.88;p=0.008), miR-1288 (AUC=0.93;p=0.027), miR-548ar-3p (AUC=0.85;p=0.009), Let-7e-5p (AUC=0.52;p=0.005) and miR-125a-5p (AUC=0.52;p=0.008) were upregulated, whereas miR-4676 (AUC=0.91;p=0.003) was downregulated, in patients with multiple stroke and/or TIA compared to single stroke in second microarray analysis (fig2a). qPCR validation results confirmed the microarray results for all studied miRNAs (p&lt;0.0001), which showed predictive value for risk of second stroke and/or TIA in ROC analysis (p&lt;0.0001-fig2b/c). Enrichment analysis: IL-2, lipid metabolism, BDNF/MAPK signaling pathway, Intellectual Disability and Alzheimer9s Disease are significantly related to stroke (fig2d). <b>Conclusions:</b> None of the identified miRNAs were studied in stroke before. All studied miRNAs have diagnostic value for acute IS. Moreover, our results show predictive significance for the risk of a second stroke. Found miRNAs and pathways have thus potential diagnostic, prognostic and therapeutic utility in neurovascular diseases. M.P. was supported financially as part of the research grant ‘OPUS’ from National Science Center, Poland (grant number 2018/31/B/NZ7/01137), and M.M and D.K. were supported by NN from the National Science Center, Poland (MEiN/2022/DIR/3261).