ABSTRACT First-episode psychosis (FEP) represents a critical stage of illness during which therapeutic interventions are believed to effectively influence long-term outcomes. Nonetheless, it is not clear what biological changes in the brain underlie FEP and subsequent disease progression. Moreover, there are no established molecular biomarkers that reflect psychosis and/or predict their longitudinal outcomes. Extracellular vesicles (EVs) are cell-derived microvesicles that contain various cellular components, such as nucleic acids, proteins, and metabolites, from the donor cells and, by fusing with other cells, transfer these components between cells in a paracrine and endocrine manner. An increasing body of evidence shows that EVs in the circulation may be useful in detecting various brain disorders at early stage and predicting their clinical outcomes. In this study, our goal is to evaluate the utility of EVs in the peripheral blood as molecular biomarkers to predict FEP and their subsequent progression. Our preliminary studies with a small cohort of FEP patients and controls suggested that microRNAs (miRNAs) were differentially expressed in plasma EVs derived from FEP patients. In parallel, our analysis of longitudinal changes in clinical/neurocognitive data of FEP patients identified potential subgroups. In the proposed study, we will test our hypothesis that peripheral blood EV-associated miRNAs (EV-miRNAs) will display changes reflecting clinical/neurocognitive data in FEP patients and their subgroups. We will first extend our preliminary findings on differential plasma EV-miRNA expression in FEP patients in a larger size of FEP patient and control samples, and then compare their profiles with those from CSF EV (Aim 1). We will also examine the utility of EV-miRNAs to distinguish FEP patient subgroups by combining unsupervised clustering of FEP patients based on longitudinal clinical/neurocognitive data changes and their plasma EV-miRNA data (Aim 2). This study will generate an unprecedented dataset of altered EV- miRNAs in the body fluids of FEP patients whose multimodal data are readily available. The findings will create a foundation for future studies to determine the utility of EV-miRNAs to predict long-term consequences of disease and potentially evaluate responsiveness to therapeutic interventions.