n recent years, research has made progress in attempting to foresee the development of psychosis. Several definitions of clinical syndromes and/or familial susceptibility have been proposed to predict transition to psychosis. Among subjects identified as being at risk, it is expected that almost a third of them will eventually transition to psychosis. Although these definitions clearly represent sensitive tools, they lack specificity and rely on clinical judgment. Hence, biomarkers are being investigated as objective biological assessment tools to refine the prediction of psychosis. This scoping review focuses on the biomarkers currently under investigation. We reviewed the studies published since January 2015 that put forth a biomarker associated with the risk of developing schizophrenia (SZ) or a SZ spectrum disorder. A search of the MEDLINE (via PubMed), EMBASE, PsycINFO (via OVID) and Cochrane Central Register of Controlled Trials electronic databases was conducted and 96 studies that investigated predictive biomarkers in the at-risk population were included. More precisely, we reviewed these studies focusing on study design, the effect size of the association between biomarkers and psychosis, and evidence for replication. The review revealed several biomarkers, including reduced duration mismatch negativity, reduced 40 Hz auditory steady-state responses and impaired gyrification parameters. In this paper, we discuss the lack of clinical use of these potential tools and the criteria that biomarkers must meet to be considered as medical guidelines in psychiatry.