3/5- Biomarkers to Enhance Early Schizophrenia Treatment (BEEST) - PROJECT SUMMARY The onset of a first episode of psychosis (FEP) in late adolescence or early adulthood often leads to lifelong disability. Timing and precision of treatment are of the essence during this critical developmental period. Unfortunately, FEP patients who do not respond to a conventional first-line antipsychotic (FL-AP) are often delayed in transitioning to clozapine (CLZ) - or never switch at all - despite the clear superiority of CLZ to FL- APs in treatment resistant individuals. However, CLZ treatment involves risks of severe side effects, including agranulocytosis and weight gain. Currently, clinicians and patients currently have no objective, clinically validated tools to guide this complex decision making in FEP. Our collaborative group has recently published work showing that a functional brain scan can help predict which FEP patients might not respond to FL-APs, such as aripiprazole and risperidone. Further, we have shown that a simple genetics test can help predict who is less likely to gain significant weight, and, similarly, who is less likely to develop agranulocytosis. We propose to conduct a multi-center, harmonized, randomized clinical trial with the goal of testing whether the use of biomarkers can lead to better outcomes for FEP patients. The goal of the proposed study is to develop a clozapine decision support tool based on these biomarkers. First, we will characterize 410 people with an FEP using three specific biomarkers: a resting state fMRI scan from which we will derive the striatial connectivity index (SCI) and two genetics tests (one for weight gain and the other for agranulocytosis). Those patients who are predicted to not respond to FL-APs, and who also have low risk of weight gain and agranulocytosis (approximate n=180), will be randomized in a triple-blind controlled study to either clozapine or an FL-AP (either aripiprazole or risperidone) for 12 weeks of treatment. Our main outcomes relate to clinical response, including positive symptoms, suicidal thinking, and days of hospitalization. We will also perform an MRI at study end to determine whether functional patterns in the brain distinguish CLZ responders from non-responders (target engagement). Critically, we are partnering with people with lived experience of psychosis and family members to help guide us during this trial, and to inform the study design and outcomes; information and choice are amongst the strongest elements of a successful therapeutic relationship. Overall, our study will evaluate the efficacy of whether using three biomarkers at the beginning of a first psychotic episode can lead to better patient outcomes for patients at risk for poor response, by rapidly introducing CLZ rather than waiting for multiple failures of FL-APs. Our key deliverable would be a clozapine decision support tool, consisting of the three biomarkers combined with our CLZ dosing strategy for FEP. Such a tool would be a necessary step in the development of precision psychiatry; if this efficacy trial is successful, a future study would then utilize implementation science to optimize strategies for dissemination of the decision support tool.